Movie Recommendation System Using User-Based Collaborative Filtering (ubcf)

Student: Abdulmalik Kaka Musa (Project, 2025)
Department of Computer Information and Communication Science
Federal University, Dutsin-Ma, Katsina State

Abstract

This project presents the development of a movie recommendation system using User-Based Collaborative Filtering (UBCF). The system is designed to alleviate the problem of information overload and enhance user satisfaction on movie streaming platforms. Using the MovieLens 100K dataset, the system identifies user preferences through rating patterns and recommends movies based on similarities computed with Pearson Correlation. The project was implemented in Python using libraries such as Pandas and Scikit-learn. Evaluation metrics such as Precision (0.82), Recall (0.75), and F1-Score (0.78) demonstrate the effectiveness of the system. Though limited by cold-start and sparsity issues, the results show promising potential for lightweight recommendation systems in real-world applications.

Keywords
movie recommendation system userbased collaborative filtering
Filters
Institutions